De‐prosecution and death: A synthetic control analysis of the impact of de‐prosecution on homicides

• Published date: 01 August 2022
• Author: Thomas P. Hogan
• Date: 01 August 2022
• DOI: http://doi.org/10.1111/1745-9133.12597

DOI: 10.1111/1745-9133 .12597

ORIGINAL ARTICLE

De-prosecution and death: A synthetic control

analysis of the impact of de-prosecution on

homicides

Thomas P. Hogan

Correspondence

Thomas P.Hogan, Gaza & Hogan LLC,

1600 Paoli Pike, Suite 300, Malvern, PA

19355, USA.

Email: tomhogan002@gmail.com

Abstract

Research Summary: De-prosecution is a policy not

to prosecute certain criminal offenses, regardless of

whether the crimes were committed. The research ques-

tion here is whether the application of a de-prosecution

policy has an effect on the number of homicides for large

cities in the United States. Philadelphia presents a natu-

ral experiment to examine this question. During 2010–

2014, the Philadelphia District Attorney’s Office main-

tained a consistent and robust number of prosecutions

and sentencings. During 2015–2019, the office engaged in

a systematic policy of de-prosecution for both felony and

misdemeanor cases. The city recorded the fewest num-

ber of criminal prosecutions in modern history, with a

70% reduction in the number of criminal sentencings.

Philadelphia experienced a concurrent and historically

large increase in homicides. This article employs a

difference-in-differences analysis using a synthetic con-

trol method to estimate the effects of de-prosecution on

the number of homicides in Philadelphia. The poten-

tial donor pool is composed of the prosecutors’ offices

for the 100 largest cities in the United States over a

10-year period, with a quantitative categorization of

the prosecutors’ offices used both as a variable and to

exclude cities that may have been subject to a similar

de-prosecution treatment. The synthetic control model

Criminology & Public Policy. 2022;21:489–534. © 2022 American Society of Criminology. 489wileyonlinelibrary.com/journal/capp

490 HOGAN

estimates that de-prosecution has been associated with

a statistically significant increase of 74.79 homicides per

year in Philadelphia during 2015–2019.

Policy Implications: As various prosecution policies

such as de-prosecution are being implemented across

the United States, such policies should be tested for

downstream results. The broadscale de-prosecution pol-

icy of Philadelphia—particularly for firearm and drug

trafficking offenses—appears to have a causal associ-

ation with a large increase in homicides. The public

in Philadelphia will have to make a normative choice

between a reduction in the number of prosecutions and

an increase in homicides. The government of Philadel-

phia may consider whether significantly decreased pros-

ecutions by the district attorney’s office should result

in a decrease in the budget for those services. The

overall relationship between de-prosecution and homi-

cides should be reviewed by prosecutors across the

nation for consideration in exercising their prosecuto-

rial discretion, given unique local considerations in each

jurisdiction.

KEYWORDS

de-prosecution, homicide, prosecution, sentencing, synthetic

control

What would happen to the number of homicides in a city if the prosecutor’s office stopped

charging and convicting defendants of a broad range of crimes? De-prosecution is the discre-

tionary decision not to prosecute certain criminal conduct, regardless of whether the crimes

actually were committed. De-prosecution can take place at multiple points in the criminal jus-

tice system: pre-offense, charging, pretrial proceedings, or sentencings. The city of Philadelphia

presents a natural experiment to answer the question about the impact of de-prosecution. Fol-

lowing a lengthy traditional period of prosecutions, the Philadelphia District Attorney’s Office

(the “DAO”) engaged in a systematic policy of de-prosecution from 2015 to 2019. This policy

covered both misdemeanors and felonies, with a particular emphasis on de-prosecuting drug pos-

session, drug trafficking, and felons possessing firearms. In this study, I evaluate whether the

policy of de-prosecution has a material impact on homicides in Philadelphia.

This article uses a two-step approach to examine the impact of de-prosecution on homicides.

First, I review descriptive statistics for prosecutorial and homicide trends in Philadelphia dur-

ing 2010–2019. From 2010 to 2014, Philadelphia employed relatively vigorous and consistent

prosecutorial policies. From 2015 to 2019, Philadelphia engaged in a methodical de-prosecution

strategy encompassing felonies and misdemeanors, corresponding to a substantial increase in the

HOGAN 491

number of homicides. Then, I use a difference-in-differences (“DiD”) analysis and synthetic con-

trol method to compare homicide data from 2010 to 2019 across the 100 largest cities in the

United States, using Philadelphia as the treatment city.Results from the synthetic control indicate

that Philadelphia’s de-prosecution strategy wasassociated with a statistically significant increase

of 74.79 homicides per year during the 2015–2019 period. From a statistical, mechanical, and

theoretical perspective, the substantial drop in successful prosecutions for felony weapons vio-

lations appears to be the strongest driver in this result. These results were confirmed via multiple

alternative modeling strategies and robustness tests. I conclude with a discussion of the impli-

cations for these findings by considering the trade-offs between de-prosecution and downstream

impacts on public safety, budgetary issues for multiple agencies in the criminal justice system,

and data-driven prosecutions.

The article makes three distinct contributions to the intersecting scholarship of criminology,

law, sociology, and economics. First, it defines the phenomenon of de-prosecution, including a

description of the mechanisms of this policy. Second, it is one of the first quantitative tests of

the impact of the policy of de-prosecution on homicides. Third, this article is the first study to

attempt a quantitative comparative classification of the largest prosecutors’ offices in the United

States. Such a taxonomy can serve as a starting point for multiple other types of analyses.

1 POLITICAL, PROCEDURAL, AND RESEARCH BACKGROUND

A short review of Philadelphia politics, criminal procedure, and prior research provides context

to this study. The history of the DAO for the past three decades is particularly relevant. From 1991

to 2009, Lynne Abraham was the Philadelphia District Attorney. She was known as the nation’s

“Deadliest District Attorney” for her zealous pursuit of the death penalty (Rosenberg, 1995). She

employed traditional law enforcementand prosecutorial strategies, maintaining prosecutions and

sentencings at high levels (Pilkington, 2016).

After a long tenure, Abraham was succeeded by R. Seth Williams. Williams, Philadelphia’sfirst

Black District Attorney,was elected in 2009 and began serving in 2010 (Duchneskie, 2017;WHYY

Staff Report, 2009). He ran as a moderate reformer, stressing being “smart on crime,” with an

emphasis on reducing misdemeanor prosecutions. However,he followed a general law-and-order

approach with regard to vigorous felony prosecutions (Otterbein, 2017). Under Williams’ leader-

ship and balanced approach, homicides in Philadelphia dropped to a historical low of 246 deaths in

2013 and 248 deaths in 2014. He was easily re-electedto a second term beginning in 2014. However,

in his second term, he immediately experienced political difficulties. The Philadelphia Demo-

cratic party criticized him for prosecuting Democratic elected officials (Flowers, 2015;Thompson,

2014). The Philadelphia police union feuded with him for proposing independent investigationsof

officer-involved shootings and refusing to allow police officers with credibility problems to testify

in criminal cases (Cipriano, 2017; Otterbein, 2015;Philly Law Blog, 2014; Roebuck, 2015). The pro-

gressive wing of the local Democratic party attacked him for using traditional prosecution tactics

(Gambacorta, 2016;Orso,2017;Owens,2015; Rice, 2017).

Following these events, in 2015, Williams began to change his prosecutorial policies in antic-

ipation of running for re-election again. He began to cut more criminal prosecutions and

de-criminalize a wider range of conduct. Thus, the de-prosecution policy in Philadelphia began in

2015, a clear policy shift that is captured by the data described in the following sections. Starting

in 2015, homicides also began rising in Philadelphia. In 2017, Williams was indicted and convicted

by federal authorities for a fraud scheme, and was subsequently sentenced to five years in prison